GROUP_READ
option also works for BAM file input. The clustering process is based on read sequences in alignments using the similar method as described in 5.2.1. After this step, a new bam file will be generated by replacing original read information with read group and removing redundant alignments in original bam file.